A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 14, 05:30 UTC
Add hardware plugins to installation doc (#22732)
Signed-off-by: Michael Goin <mgoin64@gmail.com> Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com> Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Michael Goin committed
Aug 12, 2025 at 20:14 UTC
e18859298d109870b22cb5b8672d1078818e268d
1 file changed
+13
docs/getting_started/installation/README.md
+13
@@ -14,3 +14,16 @@ vLLM supports the following hardware platforms:
14
- [Google TPU](google_tpu.md)
15
- [Intel Gaudi](intel_gaudi.md)
16
- [AWS Neuron](aws_neuron.md)
17
+
18
+## Hardware Plugins
19
+
20
+The backends below live **outside** the main `vllm` repository and follow the
21
+[Hardware-Pluggable RFC](../design/plugin_system.md).
22
+
23
+| Accelerator | PyPI / package | Repository |
24
+|-------------|----------------|------------|
25
+| Ascend NPU | `vllm-ascend` | <https://github.com/vllm-project/vllm-ascend> |
26
+| Intel Gaudi (HPU) | N/A, install from source | <https://github.com/vllm-project/vllm-gaudi> |
27
+| MetaX MACA GPU | N/A, install from source | <https://github.com/MetaX-MACA/vLLM-metax> |
28
+| Rebellions ATOM / REBEL NPU | `vllm-rbln` | <https://github.com/rebellions-sw/vllm-rbln> |
29
+| IBM Spyre AIU | `vllm-spyre` | <https://github.com/vllm-project/vllm-spyre> |